I am surprised that the article did not really mention why American life expectancy is relatively low. The answer to this question is well-known and is a related to a unique negative bias in the American statistics rather than older Americans dying earlier.
Put simply, Americans have an anomalously high rate of dying when they are young due to homicide and accidents, particularly vehicular. These are large statistical outliers compared to the rest of the industrialized world. This negatively alters the average life expectancy but does not say much about the life expectancy in the individual case.
If you control for just homicide and vehicular accidents, Americans have among the highest life expectancy in the world despite issues like obesity, and many Americans have negligible risk of those causes of premature death. In terms of deaths due to medical issues (cancer, heart disease, etc), Americans have the highest life expectancy in the world, which offsets the homicide and accident death rates among the young.
There used to be a saying by the people that study life expectancy statistics that the only people on Earth that live longer than Japanese women are Japanese women living in the US. Which is true and captured an essential point about the mixture of factors in life expectancy statistics.
Hmm? It did describe the major reason, and it wasn't younger people killing themselves in accidents or infant mortality:
Right away, they found our weak spot. "U.S. women have
relatively high mortality rates at the younger older ages,"
they said, which means when women hit their 55th birthdays,
for the next almost 20 years, roughly 55 to 75, they will
die more often than women in comparable countries. Americans
get more lung disease, more heart disease, more diabetes. If
Americans reach 75, they get competitive again, but that
early old age is where we lose ground. American men showed
pretty much the same weakness at roughly the same times.
They just haven't been able to explain why this discrepancy at this age exists.
Did you not see that part, or did you have some disagreement with it? I have not read the full report that it was based on yet, so I can't comment in detail: http://www.ncbi.nlm.nih.gov/books/NBK62373/
> Put simply, Americans have an anomalously high rate of dying when they are young due to homicide and accidents, particularly vehicular.
Except that, as noted in the NIH study cited in the report [1], the US has, over the past several decades, been progressively falling behind other OECD countries in life expectancy at age 50 in both genders, because of a higher mortality rate at the younger end of the 50+ scale. So, no, its just not about Americans getting killed (whether in accidents and homicides or any other way) at a young age.
As I understand it, life expectancy is also skewed by misreporting of infant deaths in other countries, which biases infant mortality stats against countries that reliably report them.
"Misreporting" is a strong term. While I'm sure a few countries (e.g. the Communist Paradise of Cuba) do misreport their statistics, many other countries get different answers via reasonable disagreement on definitions.
Consider an 5 month old fetus who completely passed through the vagina, takes one breath and dies. Is that infant mortality or miscarriage? Now make it 6 months old, but he didn't take any breaths. Or 7 months, the head (but not body) passed through the vagina, then he dies. Or 7 months, body but not head passed through vagina. If you want to draw distinctions between miscarriage and infant mortality you need to draw a more or less arbitrary line. Not every nation draws the line in the same place and the statistics are therefore incomparable.
You are right about the effect this has on the statistics. I'm just suggesting that there are far less nefarious reasons than misreporting stats.
Calling it misreporting is harsh, but the implication of not counting some babies is that they weren't really people. I think that's worth thinking about.
Also, calling a fully-born human being a fetus is both inaccurate and even more harsh than this use of the word "misreporting".
For the sake of furthering on-topic discussion, let's all agree that infant mortality is a different problem than longevity and therefore a better measure than "life expectancy from birth" would be "life expectancy from age 1" or something along those lines.
As the article (and even more, the NIH study cited in the article) shows, statistics show that the US does poorly against other developed countries if you look at life expectancy at age 50, which is obviously not skewed by differences in reporting infant mortality.
Are you sure about that? The issue isn't that people don't care if infants die, but that infant fatalities occurring quickly after birth can be declared "stillbirths", and that there aren't standards for that distinction.
Before everyone goes on about American healthcare and obesity, consider what the numbers are actually measuring.
As far as I can tell from the article, these life expectancy numbers do not compensate for suicide, homicide, premature births (some countries don't count them as births), and accidental death (as in car accidents).
I would like to see more of the same charts, but instead of just graphing life expectancy at birth, graph life expectancy from ages 1, 11, 21, 31, 41, and so on. I would compare the "Life Expectancy from Age X" graph for the U.S. against the same graphs for other countries. Then I would look at leading causes of death for each age group in the U.S. Then I would have a much more quantifiable idea of why citizens in other countries tend to live longer than U.S. citizens.
EDIT: This is important because counting thousands of premature births and having a much higher rate of death due to accidents and violence will really wreck the U.S. average. In that case, talking about how Americans eat too much red meat is mostly beside the point.
> I would like to see more of the same charts, but instead of just graphing life expectancy at birth, graph life expectancy from ages 1, 11, 21, 31, 41, and so on. I would compare the "Life Expectancy from Age X" graph for the U.S. against the same graphs for other countries.
The article itself notes:
In 2011, the National Institutes of Health issued a report that tried to make sense of it all. Right away, they found our weak spot. "U.S. women have relatively high mortality rates at the younger older ages," they said, which means when women hit their 55th birthdays, for the next almost 20 years, roughly 55 to 75, they will die more often than women in comparable countries. Americans get more lung disease, more heart disease, more diabetes. If Americans reach 75, they get competitive again, but that early old age is where we lose ground. American men showed pretty much the same weakness at roughly the same times.
And the NIH study [1] linked in the article actually has the kind of charts that you suggest should be done comparing life expectancy at various ages.
It would maybe help if instead of reciting canned arguments based assumptions generated by the title, people would actually read the linked articles.
My comment was to push back at already-posted assumptions about the U.S. healthcare system in general and obesity in particular. I have gone through the original study, but the NPR article said:
"The NIH report says that in 2002, 2003 and 2004, life expectancy in France increased by 10 months, which is a crazy pace (and I'm guessing it hasn't stayed that way)."
The likeliest explanation for a jump that large is a change in methodology, but there could be other factors (climate? there was a big heat wave in 2003). The U.S. does have a different culture and climate than these other countries.
I'm very happy to read an article that discusses the complexity of comparing life expectancies between countries, but the NPR article wasn't it and the HN comment section is just now getting to the point where people are discussing what the numbers really mean.
The podcast I linked is more along those lines as well, but it also goes into other popular measurements of national health as well, if I recall correctly.
If you're doing longevity studies in Europe, then you should get huge differences between cohorts separated by a few years due to WW2 - there are significant long-term effects (such as changes in metabolism gene expression) caused by, say, near-starvation during pregnancy or infanthood; and current longevity studies all include people who were essentially born on a battlefield as well as those who weren't due to small differences of age or location.
I believe that American diet may have a hand in this. I don't know when it happened exactly, maybe it was when the new Coke formula came in, however, since then an increasing amount of Americans have been 'corn fed'. The meat is corn fed, the breakfast cereal is made of corn, the milk is from corn fed cows, the sugar is made from corn and everything else is made from corn give or take a soy bean here or there.
Robert Lustig is the man to explain all this. He blames fructose (i.e. corn stuff) and fully expects this generation of Americans to have a shorter lifespan than previous generations could expect.
We just eat too damn much. Any chain restaurant has 1,4000 calorie 'salads' and you get served 2+ lbs in a pasta dish. That doesn't even count the appetizer or dessert that cram down your throat for the $12.99 3-course meal
"To my surprise, they found our health disadvantage 'could not be explained simply by reference to problems associated with an inefficient health care system, the lack of universal health care coverage, or large racial and socioeconomic disparities in the United States.'"
That doesn't seem to address the fact that as a nation, the US seems to be growing more and more obese with worse nutrition and less exercise. I assume the authors of the paper did some analysis based on things other than the inefficient health care system.
It might if the reasons we were dying between 55-75 were related to poor nutrition and low activity levels. My assumption (always dangerous) is that the things that kill us in those age ranges are diabetes, heart disease and stroke which are all manageable in some fashion through moderation in diet as well as exercise. Once you make it past 75, you've probably been either lucky genetically or you've been living well or some combination of the two.
Sure, look at how much assistance has increased, especially SSID payments. Plus throw in that getting back in the door post 50 is not so simple. Don't have anything to do all day? Eating is a refuge for many.
As far as I can tell, these graphs only show the average age of people who are dying in a given year. If someone survives, they won't count as an improvement! So to inflate the graphs, we just have to kill off all the really old people :)
A little research (sources at end) shows that the many commenters pointing to vehicular accidents, homicide, infant mortality, and suicide as the "reason" for lower US life expectancy are flat-out wrong:
Each year the chance of someone dying in a vehicular accident is about 1 in 10,000 in the US. Over 80 years that means a citizen has roughly a 1 / 125 chance of dying in a vehicular accident. The total reduction in life expectancy from vehicular accidents is thus well under 1.0 years. Assume, for example that such an accident cuts an average of 60 years off a person's life -- they would have lived to 80, say, but their life was cut short at just 20 years old. That works out as a 0.5 year reduction in life expectancy due to accidents.
This is without taking into account the vehicular fatality rates in other countries. When you do that for, e.g., Japan (1 in 20,000 per year), the difference in life expectancy is likely around 0.25 years.
The homicide rate in the US is about 1 in 20,000 people per year, or about half the vehicular fatality rate, so perhaps a further 0.25 year reduction, without factoring in homicides in other countries.
For infant mortality, the chance of dying at birth is a little over 1 in 200 in the US. That translates to a reduction in life expectancy of about 0.4 years, without factoring in infant mortality in other countries.
A similar analysis shows suicide rates are probably not a factor -- in fact, in some countries, such as Japan, the rate is higher than in the US, while in countries such as the UK it is comparable to the US.
So the contributions from all four of these sources is very unlikely to be more than a 1 year difference in life expectancy. It's probably quite a bit less, since I've ignored (e.g.) infant mortality in other countries. Certainly, the contribution from all four of these is nowhere near the dominant effect to explain differences in life expectancy.
A caveat: some people pointing to infant mortality are claiming that what's going on is that a large number of stillbirths simply aren't being reported in some countries outside the US, and this is skewing the results. That possibility isn't accounted for in my analysis above. I doubt it's true - in my many years living in Canada and Australia I've known quite a few people who died in car accidents, and none of whom had a stillbirth (to my knowledge). But while this suggests to me that this effect isn't large, it's of course only anecdotal evidence.
I think you mean 1/1250 in 80 years... thanks for scaring me. Edit: Uh, no, it is 1/125. Ouch. That seems awfully high.
I can only hope that's some bad statistics whereas your chances are disproportionally high while you're a teenager but if you make it past that, you're ok...
That humans continue to operate motor vehicles is a serious bug, not a feature.
I have kids who will be driving age in 10 years. If there is a self driving car widely available by then, and I suspect there will be, I will look to the heavens and praise Science.
A lot has changed since the 1950s. Correlation != causation.
Besides, race and poverty are so closely linked in the US that, even if this were true, it still doesn't say anything about the intrinsic properties of different races. Implying that it does is a racist statement.
Saying that members of some races are shorter on aberage than others may be "racist" in some sense, but it's also informative. Same with life expectancy.
Now how much of it is genetic vs lifestyle and cultural is a different story. It's not easy to tease apart genotype vs phenotype arguments or nature vs nurture. But I don't think I would take offense to ACTUALLY 90% TRUE generalizations about MY race and ethnicity, and I don't think others should, either.
That's different than saying POTENTIALY UNTRUE things based on anecdotal evidence, such as that Mexicans are usually bad at physics, that women bad at programming or that Jews are averse to taking low paid menial jobs. Disclaimer: I am Jewish. But I have no issue discussing things dispassionately like that. However I am sensitive to how they are presented, and MORE imporyantly careful ny to perpetuate harmful stereotypes that are possibly untrue or may cause people to actually suffer as a result.
Put simply, Americans have an anomalously high rate of dying when they are young due to homicide and accidents, particularly vehicular. These are large statistical outliers compared to the rest of the industrialized world. This negatively alters the average life expectancy but does not say much about the life expectancy in the individual case.
If you control for just homicide and vehicular accidents, Americans have among the highest life expectancy in the world despite issues like obesity, and many Americans have negligible risk of those causes of premature death. In terms of deaths due to medical issues (cancer, heart disease, etc), Americans have the highest life expectancy in the world, which offsets the homicide and accident death rates among the young.
There used to be a saying by the people that study life expectancy statistics that the only people on Earth that live longer than Japanese women are Japanese women living in the US. Which is true and captured an essential point about the mixture of factors in life expectancy statistics.